Multicore SoCs stay a step ahead of SoC FPGAs

by Texas Instruments, Apr. 14, 2016 – 

Executive summary

Historically, the differences between a system-on-a-chip (SoC) and a field programmable gate array (FPGA) were fairly obvious. Certainly there was overlap and they competed against each other in some applications, but by and large the two technologies followed their own paths. Now though, the creative marketing of FPGA vendors could suggest that they are on a collision course and are interchangeable. Nothing could be further from the truth.

Early in their evolution, FPGAs were perceived by design engineers as simply configurable logic gates which could be applied in mundane and often repetitive operations in low-volume systems that could not justify the greater expense of an application- specific integrated circuit (ASIC). Recently, the integration of Arm Cortex - A cores into FPGAs and compute-intense cores could lead one to believe that the paths of true multicore SoCs and these so-called FPGA SoCs had converged. A closer examination reveals that in reality they are still very far apart and that true multicore SoCs offer increasing advantages in those critical areas required by today's demanding products.

 Back